Power cooperative communication method and system based on advanced reward multi-agent reinforcement
By using an advanced reward multi-agent reinforcement learning method, the reward weights are dynamically adjusted to optimize the flight trajectory and power scheduling of UAVs, thus solving the high energy consumption problem of multi-UAV systems in complex environments and achieving efficient and low-energy data acquisition.
CN122419656APending Publication Date: 2026-07-17INFORMATION COMM COMPANY STATE GRID SHANDONG ELECTRIC POWER +1
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- INFORMATION COMM COMPANY STATE GRID SHANDONG ELECTRIC POWER
- Filing Date
- 2026-03-30
- Publication Date
- 2026-07-17
Smart Images
Figure CN122419656A_ABST
Abstract
本发明公开了一种基于进阶奖励多智能体强化的电力协同通信方法及系统，属于无线通信领域，针对现有技术在多无人机协同通信中难以处理多目标冲突、易陷入局部最优导致能耗过高的问题，构建多无人机辅助的电力物联网系统模型，将协同通信问题形式化为部分可观测随机博弈，引入多智能体近端策略优化算法，并结合五阶段进阶奖励塑形策略，在训练不同阶段动态调整奖励函数各分项的权重，引导策略从基础通信能力学习平滑过渡至能量绝对最小化目标；训练完成后，各无人机根据实时局部观测状态分散式执行，输出三维速度与发射功率控制动作。最终在复杂动态电力应急场景中能有效克服局部最优困境，在保障数据采集率与覆盖性能的同时降低系统总能耗。
Need to check novelty before this filing date? Find Prior Art